Use of Information Technologies for Precision Crop Protection

Most weed populations have been found to be distributed heterogeneously in time and space within agricultural fields. The spatial distribution of weeds has been ignored in weed management because equipment for automatic weed sampling techniques to measure the spatial variation of weeds have so far not been implemented in practice. A major step towards a practical solution for site-specific weed management is the development of precise and powerful sampling techniques to automatically and continuously determine in-field variation of crop cover and weed seedling populations. An automatic weed detection system using digital image analysis, computer-based decision making and GPS-controlled patch spraying was developed. In order to study this system in practice, bi-spectral images of cereals, sugar beets, maize and oil seed rape were taken under field conditions and analysed with an image analysis software to identify crops and weed species and classes of broad-leaved weeds, grass weeds and Galium aparine based on characteristic shape features. Site-specific herbicide application was performed using a GPS-controlled patch sprayer with three separated hydraulic circuits at the research station Dikopshof in Wesseling and on the farm of Manfred Hurtz in Nideggen-Berg.